Computer aided analysis of phonocardiogram

Abstract

In the present paper analysis of phonocardiogram (PCG) records are presented. The analysis has been carried out in both time and frequency domains with the aim of detecting certain correlations between the time and frequency domain representations of PCG. The analysis is limited to first and second heart sounds (S1 and S2) only. In the time domain analysis the moving window averaging technique is used to determine the occurrence of S1 and S2, which helps in determination of cardiac interval and absolute and relative time duration of individual S1 and S2, as well as absolute and relative duration between them. In the frequency domain, fast Fourier transform (FFT) of the complete PCG record, and short time Fourier transform (STFT) and wavelet transform of individual heart sounds have been carried out. The frequency domain analysis gives an idea about the dominant frequency components in individual records and frequency spectrum of individual heart sounds. A comparative observation on both the analyses gives some correlation between time domain and frequency domain representations of PCG.